To be effective, recommender systems should strike the right balance between personalized and unpersonalized features. Let's see how Netflix solves this problem with its multi-purpose recommender algorithms.
Constructing an ML (Machine Learning) system or data analytics model that properly generalizes to data outside of the training sample may be a challenging task. This article shows what options and techniques you may use to improve performance and generalizability of your ML models
Generative Adversarial Networks (GANs) is a novel model that creatively synthesizes concepts from game theory and Machine Learning. This blog describes GANs architecture and advantages over other generative models.
Enterprises drive optimization thru a new branch of Machine Learning called Reinforcement Learning. This is applied in ticket optimization, Autonomous driving and data centre energy optimization.